Whether we ultimately consider California a country or not is just an argument about the meaning of words

However, that doesn't mean it doesn't have an answer. The stopsign "that's just an argument about the meaning of words" is useful in cases where a genuine ambiguity about the meaning of a word has caused a discussion to be diverted from its main topic, which was something else. But here, the meaning of words is the topic (as I noted in my reply to RichardKennaway, I entered this thread exclusively to point out the official political status of Dubai), and there's no ambiguity about what the answer is. Indeed, there's not even any argument. Rather, what we have is me pointing out a certain fact, and you (and others) evidently seeking to justify ignorance of that fact. To the extent there is any argument, it's about how important the fact is, not about whether the fact is true.

The answers to semantic questions, when they exist, may not be ultimate or necessary or fundamental or even important, but they are still real.

But I'd certainly say that the US is a very noncentral example of a country

Well, I suppose that depends on your point of view. For me, the US is the central example of a country, because it's the one I live in and am a citizen of. I would in fact be curious to know what your idea of a "central" country is. One that's smaller? Okay, but smaller countries still have political subdivisions (as indeed, do U.S. states themselves), and the whole point of political subdivisions is that policies may differ among them. France has a lot more centralization of policy than the US does, but I'll still bet you that the municipal code of Paris is non-identical to the municipal code of Marseille. Spain's "autonomous communities" definitely have differing laws from each other: Catalonia, for example, has banned bullfighting, which would be unthinkable in other parts of the country. Do you doubt that one could multiply such examples at will?

Well of course the model that contains the correct the international-law technicalities is the model you'd want to use if you wanted to predict international-law technicalities. Just like if you want to predict where to find a tomato in a biology textbook, you should model it as a fruit. But if you want to know what to cook with it, you're better off modelling it as a vegetable.

"International-law technicalities" include things like what embassy you have to go to to get a visa to travel to the place. I dispute your implicit marginalization of these "technicalities", just as I would dispute any notion that a tomato's biological status as a fruit is mere pedantry. Biology is important, and so is international law.

But I didn't intend to get into an argument about that, because I didn't expect it to be controversial.